Paul Sagawa / Artur Pylak
203.901.1633 / 203.901.1634
sagawa@ / email@example.com
September 19, 2011
- Mobile patents were hot before Google’s purchase of MMI and its IPR portfolio. Since then, the tortuous legal process for sorting out the IPR mess has a dominant topic for investors. Market leaders are tactically litigating against one another in multiple jurisdictions, while patent “trolls” sue for royalties. Along the way, incremental legal decisions should be viewed with skepticism, as embargoes, injunctions and damage awards are typically stayed on appeal and the appeals process typically takes many years IF it plays to final judgment. Despite the market enthusiasm for IPR holders like Interdigital and Kodak, we believe that Google’s recent deals bring the industry into closer balance and that future bidding over patent-rich properties will be relatively muted. We also expect the internecine battles to shift from costly litigation to negotiation, with comprehensive cross-licensing agreements between the biggest players more likely than not. Meanwhile, patent trolls WILL exact a modest tax on the industry, but are unlikely to have a serious impact on either market demand or competitive balance.
- Mobile device IPR has been a contentious issue for over a decade, after Qualcomm upset the status quo of big telecom equipment players cross licensing each other and squeezing rivals. Qualcomm faced down a barrage of litigation, including fierce opposition from then dominant market leaders Nokia and Ericsson, eventually negotiating royalty bearing cross licenses with all of its major challengers. Qualcomm succeeded because: a) they had unique and immensely valuable technology; b) they weren’t trying to sell equipment, and thus didn’t need to cross-license patents; and c) early deals with Korean manufacturers set the market rate for their portfolio.
- The history of Qualcomm has several important lessons for investors vis a vis IPR battles. The first –Lurid damages and injunctions imposed by lower courts or initial ITC judgments are usually stayed pending appeal, and the appeals process can take years to play out. In particular, injunctions have a high standard – irreparable harm – that is difficult to meet on appeal. These rulings should be considered as important inputs to negotiations over royalties rather than a penalty that is likely to be paid.
- Second, the U.S. legal precedent for setting damages for patent infringement is a framework called “The Georgia Pacific Factors”. These factors focus heavily on market precedents – what has been paid in the past for these or like patents? – favoring companies that have established long-standing licensing policies, such as Qualcomm. Essentially, it is hard to avoid paying for IPR if other competitors have already agreed to pay. Once the first licensing agreements with a particular patent holder are negotiated, the remainder eventually follow suit with terms based on the initial agreements.
- Third, patent litigation is inherently one-way – the fact that you have a strong IPR portfolio is irrelevant in determining liability for infringing on someone else’s patents. However, counter suits alleging infringement in the other direction are a powerful tool to force negotiations. One key to Qualcomm’s success in extracting royalties is that it does not produce devices, and thus is far less subject to counter claims by the manufacturers with whom it is negotiating. Most of the disputes in the current fracas are two way, suggesting a greater likelihood of a negotiated cross-licensing solution.
- Finally, not all patents are equal. Many patent infringement claims are thrown out on validity challenges, asserting that the patent is preceded by “prior art”, or that the patent is “obvious” – software patents have been particularly vulnerable. We also note that the most important patents may be subject to industry standards association requirements that essential IPR be made available on “fair, reasonable and non-discriminatory” (FRAND) terms. While many frameworks for assessing the value of IPR portfolios have circulated, we believe the proven ability to generate royalty bearing cross licenses is the best indicator. On this basis, Qualcomm, Microsoft, Motorola and Nokia have been successful.
- The number of patents involved in a single smartphone or tablet run well into the thousands, a reasonable share of which are not controlled by direct market participants, and thus, are unlikely to be addressed by cross-licenses. Patent aggregators, who acquire patents from small holders or take on responsibility for monetizing them, and specialty IPR holders are an increasing presence in the space. The value of these IPR portfolios will shake out over time through litigation and negotiation. Once value is established, licensing will proceed quickly and act as a tax on devices in the market.
- At the start, Google and its partners have been the most vulnerable to infringement claims, spurring Google’s acquisition of Motorola and 2,000 patents from IBM. While these deals do not affect the trajectory of suits already filed against the Android ecosystem, they provide ammunition to pursue countersuits and bring Apple, Microsoft, Oracle and others to the negotiating table, eliminating the most damaging potential outcomes. Despite this, Android is still at an IPR disadvantage, particularly vs. Microsoft, and licensees will still likely face a 300-500bp royalty gap unless Google is able add to its IPR via further acquisitions or IPR alliances with unaffiliated patent holders like Ericsson, Alcatel-Lucent or Sony. In either case, we expect the Android platform to retain share leadership across form factors, particularly given Google’s proven ability to monetize its position via advertising.
- Apple has been an aggressor in the patent wars, but it is also vulnerable to the licensing demands of other IPR holders. Apple’s royalty bearing settlement with Nokia earlier this year was a watershed, confirming this vulnerability, but also revealing a pragmatic side to Apple’s previous stance. Qualcomm, Microsoft and others will likely be next, leaving Apple with a royalty advantage vs. Android, but a deficit vs. Microsoft and its partner Nokia. Again, this does not change our favorable opinion of Apple in this market, expecting it to retain a highly profitable #2 position, but it does highlight an unseen advantage for Microsoft as it looks to build relevance as a strong #3.
Stop, or I’ll Sue
These days in the mobile business, everyone is suing everyone. Apple is suing Google’s biggest Android licensees, Samsung, HTC, and Motorola, with suits and counter suits filed with the US International Trade Commission (ITC), four states, nine countries and four continents (Exhibit 1). At the same time, Microsoft has engaged Motorola in three states, plus the ITC. Oracle has joined the party, suing Google directly in California for infringement of Java patents it had acquired with Sun. It has yet to turn its attention to Android licensees (Exhibit 2). Kodak has sued Apple and Research in Motion. Apple and Nokia recently settled their multi-jurisdiction grudge match, with Apple agreeing to pay ongoing royalties as part of a comprehensive cross license, and Nokia likely to press Android licensees for similar considerations (Exhibit 3).
For investors the battlefield is confusing, with a steady barrage of preliminary findings, and lower court rulings that may yield headlines of embargoes, injunctions and triple damages, but inevitably result in nothing more concrete than a stay pending appeal. Given this confusion, it is probably useful to review the basics of patent infringement litigation.
- You can sue over the use of one patent or a million patents. However, companies usually restrict their litigation to a small number of patents where they are convinced that they will win. Increasing the number of patents does not seem to increase the chances of winning, although it provides a cushion in case some of them are proven invalid during the process. In any case, increasing the number of patents under litigation increases the complexity of the suit and thus, the cost of prosecuting it.
- You can sue in any jurisdiction where the patent is valid. Patent holders generally venue shop, choosing countries and courts where they believe they are most likely to have legal advantage and where the accused infringer is doing the most business. Suits can be filed in multiple jurisdictions on the same patents (although not in multiple states), raising the stakes and the costs for both parties.
- You can file a claim with the US International Trade Commission. The ITC is a US federal agency set up to protect against unfair trade practices by foreign nations. Claims filed with the ITC are explored by technical staff which issue recommendations to a politically appointed Administrative Law Judge who reviews the recommendations and holds a hearing. If the ALJ finds that a foreign product infringes upon US patents, the ITC is authorized to embargo the importation of the infringing product. Such a finding must be confirmed by the full three member commission and is subject to Presidential veto.
- You can sue any company which provides an infringing element to the end product. Usually, companies choose to sue the manufacturer of the competed product, but suits have been filed against makers of components, distributors and retailers as well. Choice of target can focus the size of the potential award/settlement – e.g. choose the product generating the largest revenues – on the potential for victory – e.g. choose the opponent least equipped to defend itself – or on the desire for vengeance – e.g. choose the company viewed as the most vexatious rival.
- Patent infringement suits are inherently one way. Even though companies served with infringement suits often respond with countersuits of their own, each of these claims are dealt with separately. I accuse you of infringement. That I also infringe against you has no bearing on the treatment of my initial claim. Both sets of claims will wend their way through the court system (Exhibit 4).
- Mobile industry patent cases are almost always settled. The high cost of litigating IPR cases is ample incentive for parties to reach negotiated agreements. Typically, industry disputes stem from an intractable disagreement over the relative value of patent portfolios and the legal process serves to reveal these values as incremental court and commission rulings cast tea leaves to be interpreted.
- IPR litigation takes a LONG time. Qualcomm battled Nokia and Broadcom over four years and dozens of court hearings before eventually settling its claims with comprehensive cross licenses in 2009. Ericsson and Qualcomm engaged for nearly three years until Ericsson agreed to acquire Qualcomm’s money losing equipment business as part of a cross license deal. Apple’s suit against Microsoft’s Windows product lasted 6 years until Apple’s appeal to the U.S. Supreme Court was denied, with subsidiary claims settled 3 years after. By the time patent claims are finally settled by the court, the products in question may well be obsolete and the question of future royalties may be moot (Exhibit 5).
- Patents can be thrown out in court. The patent office has an extraordinarily difficult job reviewing the millions of patent applications it receives each year. As a result, patents are often invalidated due to evidence of “prior art” – similar patents granted earlier – or “obviousness” – patented innovation is merely common sense and not worthy of patent protection. This has been particularly true for software patents which are often conceptual rather than science based. For example, a Dutch court recently invalidated (in Holland) an Apple patent for using a “slide gesture to unlock a touch screen”, as other products with “swipe to unlock” existed previous to patent being granted. This is generally the first line of defense in any patent litigation. Note also that patents expire – 17-20 years from filing depending on the jurisdiction – and the age of the patent has a great bearing on the potential for either damages or license royalties.
- Court assessed damages rely on market established licensing rates. In the rare cases when patent suits proceed all of the way to assessing damages, the framework used in US courts is a list of 15 considerations called “The Georgia Pacific Factors”. While there are many nuances in the list, the overriding principle is to use market established licensing rates as the primary guide (Exhibit 6). This has the effect of perpetuating value for well established licensing programs.
- “Essential” patents are not necessarily more valuable. Most of the participants in the mobile industry belong to the several engineering organizations that set standards specifications for interoperability amongst devices and networks. These organizations require that members pledge to license any patents found essential to a ratified standard on a “fair, reasonable and non-discriminatory” basis. This requirement presupposes agreement to market established terms and weakens the ability to use these patents as weapons in court.
- “Implementation” patents have the most value in court. Implementation patents are those patents that improve the value of an end product but that are not unavoidable in meeting a standard. For smartphones, most of the OS, applications and peripheral functions (cameras, etc.) would fall into this category. For implementation patents, there is no FRAND requirement to fall back on and patent holders are better able to use the legal process to drive more lucrative licensing arrangements.
- Damages can be trebled for “Willful Infringement” but are subject to appeal. US law provides that any damage award based on a finding of “Willful Infringement” is tripled in magnitude as a penalty. In practice, this has been hard to prove. Even if a court finds for the plaintiff and the finding is upheld after appeals are exhausted, the imposed damages can also be appealed and are often reduced.
Essentially, we view patent litigation as an important element in the process of negotiating licensing agreements. As cases proceed through multiple venues, both sides offer evidence that is used both for the court’s purposes, but also by the companies in conflict to reassess the strength of their relative negotiating positions. As such, incremental legal setbacks, such as invalidation of specific patent claims or imposition of temporary import restrictions, are more indicators of the flow of royalties in future cross licensing agreements than of ultimate court decisions.
The Story of Q
Qualcomm’s role in the history of mobile industry patent conflicts is instructive. In the early days of cellular, multiple technical standards developed in parallel, with regional champions taking the lead in their home territories. In North America, the champions were Motorola and AT&T (later Lucent), with Nortel joining the party along the way. In Europe, the leaders were Siemens, Alcatel, and Ericsson, with upstart Nokia riding Nordic enthusiasm for the emerging technology to a seat at the table. Japan was an island to itself, with Fujitsu and NEC the leading telecom suppliers.
For the most part, these companies were old, long established competitors, who stayed at home and operated under comprehensive cross-licenses of each other’s patents. This culture was, in part, a legacy of world communications technology leader AT&T’s decades anti-trust battle and a gentlemen’s agreement to freely cross license with the existing national champions but to demand substantial royalties from new companies without intellectual property to share.
By 1990, mobile operators were considering a shift to digital wireless technology. In Europe, the decision was dictated by EU industrial policy which moved to mandate a single standard for the continent. Here, upstart Nokia and Swedish stalwart Ericsson played European politics to outmaneuver French Alcatel and German Siemens which had expected to sweep through a jointly developed technology proposal. Instead, the adopted standard was chock-a-block with Nordic IPR, setting Nokia and Ericsson on their way to industry leadership, as the monolithic European market gave the fledgling GSM standard immediate scale advantage.
Meanwhile, US industrial policy was far less heavy handed, with wireless carriers free to choose amongst three separate standards – GSM, an American version called TDMA and a bleeding edge technology called CDMA promoted by a iconoclastic San Diego company called Qualcomm. A few big carriers – Verizon (then called Bell Atlantic), GTE (later merged into Verizon), Pacific Bell (later renamed AirTouch and then acquired by Verizon), and Sprint – adopted CDMA. The South Korean government, seeing an opportunity to use CDMA as a tool to vault its technology industry into the global market, also selected the standard. The politics under which these decisions were made gave Qualcomm an opening and it quickly devised a comprehensive patent licensing program to enable devices and equipment markets to develop. These early licenses, at 4% or more of device value, established the value of Qualcomm’s patent portfolio from a legal perspective.
Moving ahead, Qualcomm’s CDMA technology proved to be superior to the underlying approach used in GSM and TDMA, and as the industry looked to design a third generation standard, it adopted Qualcomm’s core IPR into the 3G specification. Qualcomm demanded the same 4% royalties on the new 3G products that it was receiving for 2G CDMA devices. Device makers, and in particular the GSM players who had not participated in CDMA previously, screamed bloody murder and litigation ensued. From 1997 to 2007, Qualcomm was fighting legal battles with as many as 10 companies at a time, including multi-year, multi-venue showdowns with Nokia, Ericsson and Broadcom (Exhibit 7).
Over the course of the decade, Qualcomm stock was buffeted by incremental legal findings in this myriad of cases. In the end, all of the cases were settled through negotiation, most resulting in cross licenses affirming royalty payments to Qualcomm and one, the Broadcom case, resulting in a one-time payment back to Broadcom for the use of its intellectual property in Qualcomm’s chipsets but affirming that Broadcom could not pass its license to Qualcomm’s IPR to its own chipset customers. In the end, we believe these positive results rest on three factors: 1) The validity and quality of Qualcomm’s intellectual property for 3G was unassailable; 2) Qualcomm had been scrupulous about filing patents for its innovations in all relevant jurisdictions; and 3)The value of Qualcomm’s IPR had been well established through its licensing program;
Keeping Emotions Out of it
While it may seem inappropriate to assign emotions to companies, pride has seemingly played a role in the willingness of some companies to continue with litigation despite its immense cost in both legal expenses and management distraction. Ericsson and Nokia remained committed to their court fights with Qualcomm long after preliminary court findings began to suggest that their challenges would be fruitless. The same can be true for Apple, which pursued its infringement suit against the “look and feel” of Microsoft’s Windows OS for 9 years, all the way to a rejection by the U.S. Supreme Court.
However, despite these occasional counterexamples, most patent holders view IPR litigation as a necessary evil in pursuit of negotiated settlements. The current market circumstances are reminiscent of Qualcomm’s battles in that certain combatants have flashed elements of emotional behavior, but the reality of the situation does seem to be sinking in.
For example, consider the contrasting approaches of Microsoft and Apple. Microsoft, having learned the importance of patents over its years in the PC business, has been diligent about filing its innovations and has amassed a wealth of software patents, many directly relevant to all of the products in the sector. Microsoft has already monetized its portfolio, on undisclosed terms, with Samsung and HTC, setting the baseline for negotiations with Motorola, which is almost certain to settle as its deal with Google closes.
On the other hand, Apple remains in a costly multi-venue, multi-product legal war with Samsung, HTC and Motorola, and has shown no signs of interest in negotiations (Exhibit 8). According to several observers, Apple’s IPR claims are not as broad or valuable as Microsoft’s and it is more vulnerable to counter suits on its own hardware imports, yet it has been far more recalcitrant. Given the unusual influence of Steve Jobs’ personality on Apple culture, it would seem plausible that Apple’s hard line boils down to an emotional response to products that have apparently used Apple’s own patented user interface innovations to wrest market share leadership in smartphones. While initial court responses have largely been in Apple’s favor, we remain convinced that the company will not eschew negotiations indefinitely. First, Google’s acquisition of Motorola and a slug of IBM patents, significantly buttresses its previous inability to strike back with countersuits, with the potential that Google with further augment its holdings in future deals. Second, Steve Jobs personal situation may reduce his influence over the IPR policy. Third, Apple’s recent licensing agreement with royalties due Nokia was a significant capitulation vs. its prior stance, and perhaps indicative of an overall softening.
98 Pound Weakling?
On the mobile patent battle field, Google has been conspicuous in its weakness. Unlike Microsoft, or even Apple, it had not grown up in a litigious environment and had not internalized the importance of amassing patents. Moreover, its on-line core business puts most of its IPR in the cloud rather than on the device, leaving it naked on Android and in no position to aid in the defense of its licensees. Over the past year, as Apple, Nokia, Oracle and others began filing claims against the Android ecosystem, Google became acutely aware of its vulnerability. The only way to fill the void was to buy mobile relevant patents.
The first parlay was a well publicized play for Nortel’s patents, which were to be auctioned out of receivership. In what can only be considered an embarrassment, Google was outbid by a consortium comprising strange bedfellows Apple, Microsoft and RIM. This was quickly followed by the acquisition of Motorola and its 17,000 issued patents and 7,000 pending applications, along with purchases of more than 2,000 patents from IPR old hand IBM. Google then followed by reassigning some of its patents to its license partner HTC for use in its defense against Apple’s suits (Exhibit 9).
While Google is dramatically better armed than it had been before the deals, it is still at an IPR disadvantage against its largest rivals. While the quality of the patents acquired from IBM is not clear, Motorola has not shown any silver bullets in its IPR battle with Apple. Most of Motorola’s best patents relate to the core workings of a wireless system and are considered essential to the major wireless standards. Unfortunately, most of these must be licensed to all comers on a FRAND basis. While Motorola has appropriately established FRAND royalty bearing licenses with most handset makers, it is difficult to use these patents as a legal weapon. We note that any of the analyses that have attempted to rank various companies’ patent holdings have focused on IPR that has been asserted in court. While this is a reasonable shortcut to the herculean task of sorting through the tens of thousands of patents that have not been asserted, it tends to short change the companies that have not been particularly litigious as of late, such as Qualcomm, Microsoft, IBM, Ericsson and Alcatel-Lucent, to the benefit of those who have been active in court.
As it stands, Google and its partners will likely owe ongoing royalties to Microsoft, Nokia, Apple and Oracle, with only modest mitigation from the newly acquired patents. It is not likely that the cumulative royalties would be prohibitive – the nature of the damages setting mechanisms in the legal system ties back to established market rates for similar intellectual property. It is rumored that Microsoft is receiving between $5 and 20 per unit on Android smartphones from Samsung and HTC, likely in the range of 2-4% of the wholesale value of the products, and less than the royalty taken by Qualcomm (Exhibit 10).
Arguably, Apple’s portfolio is less extensive and should yield lower licensing fees, while Oracle’s 7 Java patents would be worth dramatically less still. We also note that Microsoft Mobile Windows and Apple iOS phones would also be liable for royalty payments to industry IPR holders, but that the overall burden would likely be 3-5 percentage points lower than for Android products, and that some Android licensees, such as Samsung and Sony could further reduce that gap through cross licenses of their own patents.
What Google really needs is a small handful of very powerful implementation patents that it can assert against Apple and Microsoft to force détente and good faith cross licensing negotiations. One possible route for Google and its partners would be to make further IPR oriented acquisitions in an attempt to drive more advantageous cross-licensing terms. Rumors continue to swirl around IPR holders like Eastman Kodak, Research in Motion, HP’s foundering Palm unit and Interdigital. The acquisition of any of these companies would be problematic. Kodak, RIM and Palm have no value to Google beyond their patents, and for the latter two, it would face obvious anti-trust objections. These companies are also highly unlikely to consider offers to buy IPR without the business. Interdigital’s patent portfolio suffers from the same limitations as Motorola – most of its IPR is “essential” and subject to FRAND requirements and existing cross licenses. As such, we would not bet on such big deals happening.
Another option is partnering with other companies with broad IPR portfolios, securing rights to use high value implementation patents in negotiations with Apple. Google has already acquired more than 2,000 patents from IBM, which has long been the world leader in patent licensing. While the quality of these patents as ammunition is not publically known, we assume that due diligence was executed and that they are useful to the Android Ecosystem legal battles. Similar accommodations could be reached with companies like Ericsson, Sony, and Alcatel-Lucent, all of which are known to have rich troves of accumulated intellectual property and interest in blunting Apple’s potential domination of the device industry. We highlight that patents need not be sold outright, but rights could be assigned that would enable their use in litigation and cross licensing negotiations.
Finally, small companies and individuals can hold patents as well, and some of these can be immensely valuable. Already, independent firms, such as RPX, Intellectual Ventures, Allied Security Trust and Acacia technologies have gained rights to thousands of patents and begun selling licenses indemnifying the holder from litigation. These patent aggregators are a big part of a dramatic expansion of IPR litigation launched by patent holders outside the industry, also known as non-practicing entities (NPEs) (Exhibit 11). Google could pursue a similar strategy to these firms and acquire small patent holdings in search of the handful of “silver bullets” that it needs. It could also approach these firms directly and offer to purchase key patents at a premium. These strategies could prove less expensive than paying up for royalty bearing cross licenses.
Winners and Losers
We do not believe that the mobile patent wars will disrupt the overall direction of the mobile device market. While Google’s market leading Android platform is clearly at a disadvantage on IPR, we are firm in our conviction that the globe-spanning litigation against it will eventually resolve to a series of cross licenses with royalties due the more patent rich Microsoft, Apple and Oracle. While these royalties may put Android products from licensees like Samsung and HTC at a 3-5% disadvantage vs. comparable Apple and Microsoft products, these costs are bearable given the scale of the Android market and Google’s commitment to cost-free licensing. Google’s ongoing efforts to acquire IPR rights could ameliorate this advantage somewhat, albeit with one-time outlays for acquisition. While Google and its partners are clearly relative losers in this scenario, we believe that impact will be small relative to the broader opportunity for device sales, service revenue and advertising. Within the Android ecosystem, the patent wars will likely end up advantaging relatively IPR rich licensees like Samsung and Sony-Ericsson, who will likely negotiate lower royalties than the less well endowed HTC and LG (Exhibit 12).
Apple may see its litigation reign of terror blunted somewhat by Google’s more aggressive defensive stance, but it retains the upper hand. We do not believe that a long term scorched earth negotiating strategy vs. Android is in Apple’s best interests, given the high costs of litigation, the possibility of antitrust scrutiny and Apple’s own vulnerability to the IPR claims of others. We expect this war to end with a press release announcing a royalty bearing cross license with Google and its licensees at undisclosed terms. This would be good for Apple in establishing a permanent cost advantage, which we expect would support the continuance of Apple’s extraordinary device margins.
Microsoft is a clear winner. It has quietly established the value of its intellectual property without excessive litigation costs. As the patent wars settle down, we would expect Microsoft to draw royalties from all smartphones and tablets, including Apple products. Microsoft’s IPR cover will allow it to continue drawing licensing fees for its Mobile Windows platform without putting that platform at cost disadvantage vs. Android or Apple. By association, Nokia is also a clear winner. It has already negotiated terms with Apple that will have it drawing royalties on all iPhones and iPads, setting the stage for further negotiations with Android manufacturers – although these players may have already negotiated cross licenses with Nokia.
While the increased importance of patents is obviously a boon for outsider IPR holders like Interdigital, HP, RIM and Kodak, we believe the likelihood of acquisition of any of these firms is small and the realities of the IPR litigation timetable and pace of negotiations puts any payoff far in the future.